ABSTRACT

Objective.

To estimate the early impact of coronavirus disease 2019 (COVID-19) vaccination on cases in older populations in four countries (Chile, Colombia, Guatemala, and the United States of America), and on deaths in Chile and Guatemala.

Methods.

Data were obtained from national databases of confirmed COVID-19 cases and deaths and vaccinations between 1 July 2020 and 31 August 2021. In each country, pre- and post-vaccination incidence ratios were calculated for COVID-19 cases and deaths in prioritized groups (50–59, 60–69, and ≥70 years) compared with those in the reference group (<50 years). Vaccination effect was calculated as the percentage change in incidence ratios between pre- and post-vaccination periods.

Results.

The ratio of COVID-19 cases in those aged ≥50 years to those aged <50 years decreased significantly after vaccine implementation by 9.8% (95% CI: 9.5 to 10.1%) in Chile, 22.5% (95% CI: 22.0 to 23.1%) in Colombia, 20.8% (95% CI: 20.6 to 21.1%) in Guatemala, and 7.8% (95% CI: 7.6 to 7.9%) in the USA. Reductions in the ratio were highest in adults aged ≥70 years. The effect of vaccination on deaths, with time lags incorporated, was highest in the age group ≥70 years in both Chile and Guatemala: 14.4% (95% CI: 11.4 to 17.4%) and 37.3% (95% CI: 30.9 to 43.7%), respectively.

Conclusions.

COVID-19 vaccination significantly reduced morbidity in the early post-vaccination period in targeted groups. In the context of a global pandemic with limited vaccine availability, prioritization strategies are important to reduce the burden of disease in high-risk age groups.

Keywords
Vaccination; COVID-19 vaccines; adults; morbidity; Americas

RESUMEN

Objetivo.

Estimar el impacto temprano sobre los casos de enfermedad por coronavirus 2019 (COVID-19) obtenido con la vacunación contra la COVID-19 en los grupos poblacionales de edad avanzada en cuatro países (Chile, Colombia, Estados Unidos de América y Guatemala), así como el efecto en la mortalidad en Chile y Guatemala.

Métodos.

Los datos se obtuvieron a partir de las bases de datos nacionales sobre vacunaciones y sobre casos de COVID-19 y muertes debidas a esta enfermedad entre el 1 de julio del 2020 y el 31 de agosto del 2021. Para cada país, se calcularon las razones de incidencia de casos de COVID-19 y de muertes por COVID-19 anteriores y posteriores a la vacunación en los grupos priorizados (50-59, 60-69 y ≥70 años) en comparación con las del grupo de referencia (<50 años). Se calculó el efecto de la vacunación expresado en forma de variación porcentual de la razón de las incidencias entre el período anterior y el posterior a la vacunación.

Resultados.

Tras la introducción de la vacuna, la razón de los casos de COVID-19 entre las personas ≥50 años y las <50 disminuyó significativamente en un 9,8% (IC del 95%: 9,5% a 10,1%) en Chile, en un 22,5% (IC del 95%: 22,0% a 23,1%) en Colombia, en un 7,8% (IC del 95%: 7,6% a 7,9%) en Estados Unidos de América y en un 20,8% (IC del 95%: 20,6% a 21,1%) en Guatemala. Las reducciones de la razón fueron máximas en las personas adultas ≥70 años. El efecto de la vacunación sobre las muertes, una vez incorporados los desfases cronológicos, fue máximo en el grupo de personas ≥70 años, tanto en Chile como en Guatemala: 14,4% (IC 95%: 11,4% a 17,4%) y 37,3% (IC 95%: 30,9% a 43,7%), respectivamente.

Conclusiones.

La vacunación contra la COVID-19 redujo significativamente la morbilidad en el período inmediato posterior a la vacunación en los grupos destinatarios. En el contexto de una pandemia con disponibilidad limitada de vacunas a nivel mundial, las estrategias de asignación de prioridades son un factor importante para reducir la carga de morbilidad en los grupos etarios de alto riesgo.

Palabras clave
Vacunación; vacunas contra la COVID-19; adulto; morbilidad; Américas

RESUMO

Objetivo.

Estimar o impacto inicial da vacinação contra a doença pelo coronavírus 2019 (COVID-19) nos casos em populações idosas de quatro países (Chile, Colômbia, Guatemala e Estados Unidos da América) e nas mortes no Chile e na Guatemala.

Métodos.

Os dados foram obtidos de bancos de dados nacionais de casos e mortes confirmados por COVID-19 e de vacinações entre 1º de julho de 2020 e 31 de agosto de 2021. Em cada país, foram calculadas taxas de incidência pré e pós-vacinação de casos e mortes por COVID-19 em grupos priorizados (50 a 59, 60 a 69 e ≥70 anos) em comparação com o grupo de referência (<50 anos). O efeito da vacinação foi calculado como a mudança percentual nas taxas de incidência entre os períodos pré e pós-vacinação.

Resultados.

A incidência de casos de COVID-19 em pessoas com idade ≥50 anos em relação às com idade <50 anos diminuiu significativamente após a implementação da vacina, em 9,8% (IC 95%: 9,5 a 10,1%) no Chile, 22,5% (IC 95%: 22,0 a 23,1%) na Colômbia, 20,8% (IC 95%: 20,6 a 21,1%) na Guatemala e 7,8% (IC 95%: 7,6 a 7,9%) nos EUA. As reduções na incidência foram maiores em adultos com idade ≥70 anos. O efeito da vacinação sobre as mortes, com defasagens temporais incorporadas, foi maior na faixa etária ≥70 anos no Chile e na Guatemala, 14,4% (IC de 95%: 11,4 a 17,4%) e 37,3% (IC de 95%: 30,9 a 43,7%), respectivamente.

Conclusões.

A vacinação contra a COVID-19 reduziu significativamente a morbidade no início do período pós-vacinação nos grupos-alvo. No contexto de uma pandemia mundial com disponibilidade limitada de vacinas, estratégias de priorização são importantes para reduzir a carga de doença em grupos etários de alto risco.

Palavras-chave
Vacinação; vacinas contra COVID-19; adulto; morbidade; América

The coronavirus disease 2019 (COVID-19) pandemic has had unprecedented effects on human health. By January 22 2023, 664 million COVID-19 cases and 6.7 million deaths had been reported worldwide (¹1. World Health Organization. WHO coronavirus (COVID-19) dashboard [Internet]. Geneva: WHO [cited 2023 Jan 31]. Available from: https://covid19.who.int
https://covid19.who.int... ). As part of the recommended control measures, vaccination against COVID-19 began in December 2020 (²2. Basta NE, Moodie EMM on behalf of the VIPER (Vaccines, Infectious disease Prevention, and Epidemiology Research) Group COVID-19 Vaccine Development and Approvals Tracker Team. COVID-19 vaccine development and approvals tracker [Inter. VIPER; 2022 [cited 2023 Jan 30]. Available from: https://covid19.trackvaccines.org/vaccines/approved/
https://covid19.trackvaccines.org/vaccin... ) following phased implementation plans that prioritized high-risk populations. The main objective was to prevent severe disease and death caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Data from clinical trials and surveillance have demonstrated high efficacy and effectiveness of COVID-19 vaccines in preventing deaths and hospitalizations (³3. Olliaro P, Torreele E, Vaillant M. COVID-19 vaccine efficacy and effectiveness—the elephant (not) in the room. Lancet Microbe. 2021;2(7):e279–e280. https://doi.org/10.1016/S2666-5247(21)00069-0
https://doi.org/10.1016/S2666-5247(21)00... ), with some variations between vaccine products and circulating variants (⁴4. Institute for Health Metrics and Evaluation. COVID-19 vaccine efficacy summary [Internet]. Seattle, WA: IHME; 2022 [cited 2022 Mar 14]. Available from: https://www.healthdata.org/covid/covid-19-vaccine-efficacy-summary
https://www.healthdata.org/covid/covid-1... ). Evaluating the effect of COVID-19 vaccination on morbidity and mortality reduction at the population level is needed to understand its benefits for public health.

Because of the limited availability of COVID-19 vaccines globally, most countries implemented a phased roll-out prioritizing high-risk populations (⁵5. World Health Organization. WHO SAGE roadmap for prioritizing uses of COVID-19 vaccines: an approach to optimize the global impact of COVID-19 vaccines, based on public health goals, global and national equity, and vaccine access and coverage scenarios, first issued 20 October 2020, updated: 13 November 2020, updated: 16 July 2021, latest update: 21 January 2022. Geneva: WHO; 2022 [cited 2023 Jan 31]. Available from: https://apps.who.int/iris/handle/10665/351138
https://apps.who.int/iris/handle/10665/3... ). The speed of vaccination roll-out in the Americas region varied greatly according to countries’ access to vaccines. The United States of America (USA) and Chile achieved 10% coverage with one dose in the general population as early as February 2021, while other countries such as Colombia and Guatemala reached this coverage by mid-2021 (Supplementary material, Appendix A). In most settings in the Americas, the introduction of the COVID-19 vaccine coincided with the emergence of the Delta variant of SARS-CoV-2 in May 2021. Compared with previously circulating variants, the Delta variant was characterized by a greater reproduction number (⁶6. Hart WS, Miller E, Andrews NJ, Waight P, Maini PK, Funk S, et al. Generation time of the alpha and delta SARS-CoV-2 variants: an epidemiological analysis. Lancet Infect Dis. 2022;22(5):603–10. https://doi.org/10.1016/S1473-3099(22)00001-9
https://doi.org/10.1016/S1473-3099(22)00... ) and an increased risk of hospitalization (⁷7. World Health Organization. COVID-19 weekly epidemiological update, edition 58, 21 September 2021. Geneva: WHO; 2021 [cited 2023 Jan 31]. Available from: https://apps.who.int/iris/handle/10665/345456
https://apps.who.int/iris/handle/10665/3... ). Its relative severity by age group was similar to previously circulating strains (⁸8. Fisman DN, Tuite AR. Age-specific changes in virulence associated with SARS-CoV-2 variants of concern. Clin Infect Dis. 2022 Aug 24;75(1):e69–e75. https://doi.org/10.1093/cid/ciac174
https://doi.org/10.1093/cid/ciac174... ).

Because of epidemiological and virological changes during vaccination roll-out, measuring the effect of COVID-19 vaccines cannot be achieved using a traditional pre–post approach in a given population. As an alternative approach, studies in the USA (⁹9. McNamara LA, Wiegand RE, Burke RM, Sharma AJ, Sheppard M, Adjemian J, et al. Estimating the early impact of the US COVID-19 vaccination programme on COVID-19 cases, emergency department visits, hospital admissions, and deaths among adults aged 65 years and older: an ecological analysis of national surveillance data. Lancet. 2022;399(10320):152–60. https://doi.org/10.1016/S0140-6736(21)02226-1
https://doi.org/10.1016/S0140-6736(21)02... ) and Israel (¹⁰10. Rossman H, Shilo S, Meir T, Gorfine M, Shalit U, Segal E. COVID-19 dynamics after a national immunization program in Israel. Nat Med. 2021;27(6):1055–61. https://doi.org/10.1038/s41591-021-01337-2
https://doi.org/10.1038/s41591-021-01337... ) estimated the effect of COVID-19 vaccination on prioritized groups by evaluating relative incidence changes and temporal changes in weekly case numbers. These studies concluded that the initial roll-out of COVID-19 vaccinations was associated with significant reductions in COVID-19 cases and hospital admissions in older adults. However, estimates from the Americas region, where almost one third of global COVID-19 cases and half of global deaths were reported (¹¹11. Taylor L. Covid-19: countries in the Americas are warned not to lower their guard. BMJ. 2022;376:o664. https://doi.org/10.1136/bmj.o664
https://doi.org/10.1136/bmj.o664... ), are scarce. This study aimed to estimate the initial effect of COVID-19 vaccination roll-out on COVID-19 cases and deaths in older populations in four countries in the region of the Americas.

METHODS

An ecological study was conducted using national databases of confirmed COVID-19 cases and vaccination between 1 July 2020 and 31 August 2021. We estimated the ratios of COVID-19 cases in persons ≥50 years to COVID-19 cases in those <50 years before and after the introduction of COVID-19 vaccines in four countries in the Americas: Chile, Colombia, Guatemala, and the USA.

Data sources

The four countries included had publicly available age-specific data on daily confirmed COVID-19 cases, deaths and vaccination dates (¹²12. Centers for Disease Control and Prevention. COVID-19 case surveillance public use data [Internet]. Atlanta, GA: CDC; 2023 [cited 2023 Jan 31]. Available from: https://data.cdc.gov/Case-Surveillance/COVID-19-Case-Surveillance-Public-Use-Data/vbim-akqf
https://data.cdc.gov/Case-Surveillance/C... –¹⁶16. Ministerio de Salud Pública y Asistencia Social de Guatemala, Departamento de Epidemiología. Situación de COVID-19 en Guatemala [Internet]. Guatemala City; 2021. Spanish [cited 2023 Jan 31]. Available from: https://tablerocovid.mspas.gob.gt/
https://tablerocovid.mspas.gob.gt/... ). Age-stratified data between 1 July 2020 and 31 August 2021 were obtained from each country’s daily COVID-19 surveillance data published officially by their health ministries. No substantial changes were reported in local surveillance systems during the study period – that is, case definitions and testing policies did not change (Supplementary material, Appendix A). This analysis period was not affected by the widespread availability of self-tests for COVID-19 diagnosis that could be done at home, with no requirement for reporting to the ministry of health. Previously published population estimates by age group for each country were used (¹⁷17. Population pyramids of the world from 1950 to 2100 [Internet]. 2019. [cited 2023 Jan 31. Available from: https://www.populationpyramid.net/
https://www.populationpyramid.net/... ).

Data management

A standardized vertical database format was established to conduct the same analysis for each country. Rows recorded daily data on confirmed COVID-19 cases and deaths by age group (<50, 50–59, 60–69, and ≥70 years) and by country. The same format was used to record vaccination data by age group to calculate first dose vaccination coverage. Data transformations included summarizing and grouping the data into specific age groups used in this analysis and making format changes to create vertical datasets.

Definitions

COVID-19 cases were cases confirmed by either antigen or polymerase chain reaction testing, according to each country’s laboratory protocol. Vaccination coverage was analyzed based on at least one dose of any COVID-19 vaccine (either as part of a one- or two-dose schedule) for the following age groups: <50, 50–59, 60–69, and ≥70 years. To capture a period with stable access to COVID-19 testing, the pre-vaccination period starting point was defined as 1 July 2020. To account for the 14-day period needed to build an immune response and start to see a population effect of the vaccination program, the end of the pre-vaccination period in each country was defined as the date when each eligible group reached 10% coverage with one dose of the vaccine. The early post-vaccination period in each country was defined as the dates between reaching 10% coverage with one dose of the vaccine in each prioritized age group and reaching 10% vaccination coverage in the reference group (<50 years).

Statistical analysis

To measure the effect of vaccination on morbidity, using data from each country, we calculated the incidence ratio of confirmed COVID-19 cases in the groups prioritized for early vaccination (50–59, 60–69, and ≥70 years) to confirmed COVID-19 cases in the reference group (<50 years). These ratios were calculated for the pre- and post-vaccination periods. A difference-in-differences approach was used to calculate the effect estimate as the relative percentage change in the incidence ratios between pre- and post-vaccination periods (⁹9. McNamara LA, Wiegand RE, Burke RM, Sharma AJ, Sheppard M, Adjemian J, et al. Estimating the early impact of the US COVID-19 vaccination programme on COVID-19 cases, emergency department visits, hospital admissions, and deaths among adults aged 65 years and older: an ecological analysis of national surveillance data. Lancet. 2022;399(10320):152–60. https://doi.org/10.1016/S0140-6736(21)02226-1
https://doi.org/10.1016/S0140-6736(21)02... ). Absolute numbers of cases were used, assuming that the changes in the population denominator over the study period would be minimal. By comparing age-specific incidence ratios over time, this analysis aimed to minimize age-associated changes in the effect estimates. Confidence intervals (CIs) for the pre/post ratio percentage changes were calculated using fractional uncertainties for each ratio. To address potential pre-existing changes in the ratio, rolling 7-day averages were calculated, and pre- and post-vaccination trends were compared using the Chow test for F statistic, defining p<0.05 as statistically significant.

The same analysis was done to measure the effect of COVID-19 vaccination on mortality using data on confirmed COVID-19 deaths in Chile (¹⁴14. Ministerio de Ciencia, Tecnología, Conocimiento e Innovación de Chile. Base de Datos COVID-19 [Internet]. Santiago; 2023. Spanish [cited 2023 Jan 31]. Available from: http://www.minciencia.gob.cl/covid19/
http://www.minciencia.gob.cl/covid19/... ) and Guatemala (countries for which data were publicly available) (¹⁸18. Ministerio de Salud Pública y Asistencia Social de Guatemala, Departamento de Epidemiología. Reporte de fallecidos COVID-19 Guatemala. Guatemala City: Ministerio de Salud Pública y Asistencia Social de Guatemala; 2020.).

Two additional analyses incorporating time lags were conducted. To account for the time needed to produce a protective immune response after COVID-19 vaccination (¹⁹19. World Health Organization. Coronavirus disease (COVID-19): Vaccines [Internet]. Geneva: WHO; 2022 [cited 2023 Jan 31]. Available from: https://www.who.int/news-room/questions-and-answers/item/coronavirus-disease-(covid-19)-vaccines
https://www.who.int/news-room/questions-... ), the end of the pre- and post-vaccination periods was defined by adding 14 days to the date when 10% one-dose vaccination coverage was reached in the reference and comparison groups. To further account for the median time for progression from COVID-19 infection to death (²⁰20. Centers for Disease Control and Prevention. COVID-19 pandemic planning scenarios [Internet]. Atlanta, GA: CDC; 2021 [cited 2023 Jan 31]. Available from: https://www.cdc.gov/coronavirus/2019-ncov/hcp/planning-scenarios.html
https://www.cdc.gov/coronavirus/2019-nco... ), the end of pre- and post-vaccination periods was defined by adding 28 days to the date when 10% one-dose vaccination coverage was reached in the reference and comparison groups.

Additional sensitivity analyses using different reference age-group populations (18–49 years and 40–49 years) were also conducted.

All analyses were conducted using R statistical software, version 4.0.2 (R Foundation, Vienna Austria).

Ethics

This secondary analysis was conducted using publicly available routine surveillance data which did not have any personal identifiers; hence ethical approval was not considered necessary.

RESULTS

Vaccination effect on COVID-19 cases

Age group ≥50 years versus <50 years. The ratio of COVID-19 cases in those aged ≥50 years to those aged <50 years decreased significantly after vaccine implementation by 9.8% (95% CI: 9.5 to 10.1%) in Chile, 22.5% (95% CI: 22.0 to 23.1%) in Colombia, 20.8% (95% CI: 20.6 to 21.1%) in Guatemala, and 7.8% (95% CI: 7.6 to 7.9%) in the USA (Table 1). The trend in the ratio of COVID-19 cases between age groups ≥50 and <50 years significantly decreased after vaccine implementation in all countries (Chow test p<0.05) (Figure 1).

Prioritized age groups versus <50 years. The pre/post ratio percentage changes in COVID-19 cases by age group (50–59, 60–69, and ≥70 years versus <50 years) ranged from: 3.8% to 14.8% in Chile; 13.3% to 30.8% in Colombia; 16.0% to 23.1% in Guatemala; and 2.8% to 12.5% in the USA (Table 1). This change was highest in adults ≥70 years who were the group eligible for first vaccination, and lowest in the age group 50–59 years who were the last of these three age groups to be vaccinated. The trend in the ratio of COVID-19 cases for all three age groups ≥50 years compared with <50 years significantly decreased after vaccine implementation in Colombia, Guatemala, and the USA (Figure 2). A significantly greater decreasing trend in the ratio of COVID-19 cases was observed for age group ≥60 years versus <50 years after vaccine implementation in Chile, but this was not seen for the age group 50–59 years (Chow test p<0.05) (Figure 2).

Sensitivity analysis. Additional sensitivity analyses were conducted using age groups 18–49 and 40–49 years as the reference groups (instead of <50 years). The result showed slightly lower percentage changes in the pre- and post-vaccination ratios of COVID-19 cases in those aged ≥50 years compared with those aged 18–49 or 40–49 years in Colombia and Guatemala. When using these alternative reference groups, the effect of vaccination was close to zero in the USA and Chile, where the analysis could only be done in age groups 60–69 and ≥70 years. In the 50–59 year age group in Chile, the date when 10% coverage was reached was the same for the comparison and reference groups, hence the analysis was not feasible.

Incorporating a 14-day lag time in the pre-vaccination period resulted in higher percentage reductions in the pre/post vaccination ratios of COVID-19 cases between age groups in all countries (Supplementary material, Appendix B).

Vaccination effect on COVID-19 deaths

The percentage change in the ratio of confirmed deaths between pre- and post-vaccination periods in Chile, without time lags was –6.4% (95% CI: –5.0 to –7.8%) in adults aged ≥70 years and 4.3% (95% CI: 3.0 to 5.6%) in those aged 50–59 years (Table 2). In Guatemala, the percentage change was highest in the age group ≥70 years (35.1%; 95% CI: 28.1 to 42.1%) and lowest in the age group 50–59 years (27.5%; 95% CI: 22.3 to 32.7%). Incorporating 14- and 28-day time lags in the pre- and post-vaccination periods resulted in higher percentage changes in the ratio of COVID-19 deaths between the age groups compared in both Chile and Guatemala. The percentage change in the ratio of confirmed deaths between pre- and post-vaccination periods incorporating time lags in Chile and Guatemala was highest in the age group ≥70 years: 14.4% (95% CI: 11.4 to 17.4%) and 37.3% (95% CI: 30.9 to 43.7%), respectively.

The trend in the ratio of COVID-19 deaths for all three age groups ≥50 years compared with <50 years significantly decreased after vaccine implementation in Chile and Guatemala compared with the pre-vaccination period (Chow test p<0.05) (Figure 3).

Sensitivity analysis. Using the age groups 18–49 and 40–49 years as reference (instead of <50 years), percentage changes in the ratio of COVID-19 deaths between the age groups before and after vaccine implementation ranged from 22.1% to 35.1% in Guatemala. In Chile, where the analysis could only be done for age groups 60–69 and ≥70 years, the vaccine effect was close to zero (Supplementary material, Appendix B).

DISCUSSION

This study estimated the effect of COVID-19 vaccination in prioritized age groups in the early post-vaccination period in four countries of the Americas. These results show that the early phases of COVID-19 vaccination programs led to significant reductions in confirmed COVID-19 cases and deaths in the older adults prioritized to receive the vaccine (²¹21. Bialek S, Boundy E, Bowen V, Chow N, Cohn A, Dowling N, et al. Severe outcomes among patients with coronavirus disease 2019 (COVID-19)—United States, February 12–March 16, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(12):343–6. https://doi.org/10.15585/mmwr.mm6912e2
https://doi.org/10.15585/mmwr.mm6912e2... ).

COVID-19 morbidity

Between pre- and early post-COVID-19 vaccination periods, reductions in the ratio of confirmed COVID-19 cases in adults aged ≥50 years versus those <50 years ranged from 7.8% in the USA to 22.5% in Colombia. The effect of vaccination on confirmed cases was greatest in the oldest age group (≥70 years) in the four countries. This reduction in the ratio of cases in the age groups prioritized for vaccination versus the reference group (<50 years old) continued for about 2 weeks after 10% vaccination coverage was achieved in the reference group. After this, the ratios of prioritized age groups to the reference group increased because the gap in vaccination coverage between the groups declined.

The effect of COVID-19 vaccination in the age group ≥70 years was greater than in the age group 50–69 years in all countries, indicating a dose–response effect – that is, higher percentage reduction in the ratio of cases in groups with higher vaccination coverage and longer observation time. During the study period, and because of the prioritization strategies used, older age groups (≥70 years) had greater coverage with two doses of a COVID-19 vaccine compared with younger prioritized age groups (50–59 or 60–69 years), which increased their time to develop an adequate immune response against COVID-19. When accounting for the time to mount an adequate immune response to COVID-19 vaccination (by increasing the pre-vaccination period by 14 days), our sensitivity analysis showed a greater effect of COVID-19 vaccines on the number of confirmed cases. Given COVID-19 vaccine effectiveness is higher after receiving a second dose of the vaccine (²²22. Lopez Bernal J, Andrews N, Gower C, Robertson C, Stowe J, Tessier E, et al. Effectiveness of the Pfizer-BioNTech and Oxford-AstraZeneca vaccines on covid-19 related symptoms, hospital admissions, and mortality in older adults in England: test negative case–control study. BMJ. 2021;373:n1088. https://doi.org/10.1136/bmj.n1088
https://doi.org/10.1136/bmj.n1088... ), a greater effect would be expected as vaccination coverage with two doses increased in the studied populations. Our results are in line with another study that showed a reduction in the COVID-19 case incidence ratio in older populations in the USA when differences on vaccination coverage between adults aged ≥65 years and those aged 50–64 years were greatest (⁹9. McNamara LA, Wiegand RE, Burke RM, Sharma AJ, Sheppard M, Adjemian J, et al. Estimating the early impact of the US COVID-19 vaccination programme on COVID-19 cases, emergency department visits, hospital admissions, and deaths among adults aged 65 years and older: an ecological analysis of national surveillance data. Lancet. 2022;399(10320):152–60. https://doi.org/10.1016/S0140-6736(21)02226-1
https://doi.org/10.1016/S0140-6736(21)02... ). Our analysis provides additional results on the effect of COVID-19 vaccination in selected Latin American countries, from which estimates were previously lacking, and highlights the importance of vaccinating high-risk groups to reduce the COVID-19 burden in older adults.

COVID-19 mortality

Using COVID-19 mortality data from Chile and Guatemala, we observed that the effect of COVID-19 vaccination was proportional to the time of observation and it was greater for mortality than morbidity in older age groups, similar to findings in other studies (²³23. Suthar AB, Wang J, Seffren V, Wiegand RE, Griffing S, Zell E. Public health impact of covid-19 vaccines in the US: observational study. BMJ. 2022;377:e069317. https://doi.org/10.1136/bmj-2021-069317
https://doi.org/10.1136/bmj-2021-069317... , ²⁴24. Jabłońska K, Aballéa S, Toumi M. The real-life impact of vaccination on COVID-19 mortality in Europe and Israel. Public Health. 2021;198:230–7. https://doi.org/10.1016/j.puhe.2021.07.037
https://doi.org/10.1016/j.puhe.2021.07.0... ). Adding lag time to the observation period to account for the time required to achieve an adequate immune response (14 days) and progression from infection to death (28 days), resulted in greater effects which are closer to reality – especially seen in Chile where the effect of COVID-19 vaccination was not observed without these lag times. This finding is in line with data from Brazil, where an important decline in mortality in adults aged ≥80 years was seen after the introduction of the COVID-19 vaccine (²⁵25. Victora CG, Castro MC, Gurzenda S, Medeiros AC, França GVA, Barros AJD. Estimating the early impact of vaccination against COVID-19 on deaths among elderly people in Brazil: analyses of routinely-collected data on vaccine coverage and mortality. EClinicalMedicine. 2021;38:101036. https://doi.org/10.1016/j.eclinm.2021.101036
https://doi.org/10.1016/j.eclinm.2021.10... ). Studies in Canada, France, and Israel also showed that initial vaccination efforts directed at adults aged 65 years and older resulted in significant overall declines in COVID-19 cases and deaths, ranging from 21% to 80% in COVID-19 cases and 30% to 87% in COVID-19 deaths (¹⁰10. Rossman H, Shilo S, Meir T, Gorfine M, Shalit U, Segal E. COVID-19 dynamics after a national immunization program in Israel. Nat Med. 2021;27(6):1055–61. https://doi.org/10.1038/s41591-021-01337-2
https://doi.org/10.1038/s41591-021-01337... , ²⁶26. Sadarangani M, Abu Raya B, Conway JM, Iyaniwura SA, Falcao RC, Colijn C, et al. Importance of COVID-19 vaccine efficacy in older age groups. Vaccine. 2021;39(15):2020–3. https://doi.org/10.1016/j.vaccine.2021.03.020
https://doi.org/10.1016/j.vaccine.2021.0... , ²⁷27. Romain-Scelle N, Elias C, Vanhems P. COVID-19 vaccine is correlated with favourable epidemiological indicators in the Auvergne-Rhône-Alpes region (France): an ecological study. Vaccine. 2022;40(5):695–700. https://doi.org/10.1016/j.vaccine.2021.12.036
https://doi.org/10.1016/j.vaccine.2021.1... ). Given the greater COVID-19 vaccine efficacy against hospitalization and death, further studies to evaluate the effect of COVID-19 vaccines on different clinical outcomes, such as mortality or hospitalization, in the general population would be useful to increase the evidence of the public health benefits of COVID-19 vaccination. Furthermore, running these specific analyses by vaccine type and schedule could provide valuable insights to support vaccination programs.

Study limitations

For the approach used in our study to be valid, non-vaccination prevention and diagnostic measures and relative risks of COVID-19 between age groups had to remain constant over pre- and post-vaccination periods. The data gathered for the countries included in this analysis indicated that no changes occurred in age-group specific testing policies or behaviors over the study period (²⁸28. Ritchie H, Mathieu E, Rodés-Guirao L, Appel C, Giattino C, Ortiz-Ospina E, et al. Mexico: coronavirus pandemic country profile [Internet]. Oxford: Our World in Data; 2020 [cited 2021 Oct 11]. Available from: https://ourworldindata.org/coronavirus/country/mexico
https://ourworldindata.org/coronavirus/c... , ²⁹29. Hale T, Angrist N, Goldszmidt R, Kira B, Petherick A, Phillips T, et al. A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nat Hum Behav. 2021r;5(4):529–38. https://doi.org/10.1038/s41562-021-01079-8
https://doi.org/10.1038/s41562-021-01079... ). In addition, relative age-specific risks of symptomatic infection and severity with the Delta variant and previous variants of concern were similar (⁸8. Fisman DN, Tuite AR. Age-specific changes in virulence associated with SARS-CoV-2 variants of concern. Clin Infect Dis. 2022 Aug 24;75(1):e69–e75. https://doi.org/10.1093/cid/ciac174
https://doi.org/10.1093/cid/ciac174... ). With regard to non-pharmaceutical prevention measures, changes in requirements for physical distancing, isolation and mask use were frequent in all countries. If applied to all age groups at the same time, these measures should have had limited impact on our estimates. However, in settings where age groups have different exposures to non-pharmaceutical measures over time, the estimates could be affected. This bias could apply to countries where a high proportion of older people live in long-term care facilities, such as in European countries or the USA. In Latin American countries, where social mixing between age groups is higher, smaller differences between age groups would be expected in exposure to non-pharmaceutical measures and hence to SARS-CoV-2 (³⁰30. Mousa A, Winskill P, Watson OJ, Ratmann O, Monod M, Ajelli M, et al. Social contact patterns and implications for infectious disease transmission – a systematic review and meta-analysis of contact surveys. eLife. 2021;10:e70294. https://doi.org/10.7554/eLife.70294
https://doi.org/10.7554/eLife.70294... , ³¹31. Mossong J, Hens N, Jit M, Beutels P, Auranen K, Mikolajczyk R, et al. Social contacts and mixing patterns relevant to the spread of infectious diseases. PLoS Med. 2008;5(3):e74. https://doi.org/10.1371/journal.pmed.0050074
https://doi.org/10.1371/journal.pmed.005... ). To ensure the populations were as similar as possible, sensitivity analyses were conducted using reference age groups of 18–49 and 40–49 years. The results of these analyses were similar to the main analysis in Guatemala and Colombia, validating the reference group selection in these countries. The effect of COVID-19 vaccination observed through this analysis is directly related to the speed at which vaccination in the reference groups reached 10% vaccination coverage with at least one dose of the vaccine. In the Chile and the USA, where vaccination roll-out was faster, comparison and reference groups reached this cut-off at a similar time, resulting in a very short (or no) observation time and, consequently, a measured effect close to zero. Finally, it is important to stress that this work is based on the secondary analysis of surveillance data, for which direct validation checks were not possible to make.

CONCLUSION

Ratio analysis is a practical approach to measure the effect of phased vaccine introductions in rapidly evolving virological settings. The findings of this analysis suggest that COVID-19 vaccination had a rapid and significant effect on COVID-19 morbidity in older age groups in four countries in the Americas in the early post-vaccination period. A dose–response effect was observed, with the greatest effect in the groups that were targeted first (≥70 years). In the context of a global pandemic, where vaccine availability is limited, these findings highlight the relevance of prioritization strategies to reduce the burden of disease in high-risk age groups. Further analyses should focus on evaluating the effect on hospitalization and mortality in other countries, which were the primary targeted outcomes for the current COVID-19 vaccines. Prioritization of high-risk groups for COVID-19 vaccination should be continued to reduce virus transmission and severe outcomes of infection.

Disclaimer.

The authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/PAJPH or the Pan American Health Organization (PAHO).

Acknowledgements.

We thank the following PAHO colleagues in Washington, DC in the Comprehensive Immunization Program (previously FPL/IM) and Public Health Emergency Program (PHE) for their critical discussions on the design and their useful review of the manuscript: Silvano Barbosa, Pamela Burgos, Marcela Contreras, Tshewang Dorji, Margherita Ghiselli, Esther Kukielka, Rafael León, Lidia Redondo, Ana Riviere-Cinnamond, and Andrea Vicari.

References

Publication Dates

History